In general, the LAMP isothermal nucleic acid amplification tests work better and faster with multiple primer pairs; and I would be VERY suspicious of the specificity of such a reaction operating at low temperatures (and I'm not aware of any DNA polymerases that operate efficiently at such temperatures. LAMP generally uses the Bst polymerase at 60+ deg C. Then again, I don't know everything, and haven't been in a PCR or LAMP lab for some years now)
The reasons for using higher temperatures are related to stringency of primer binding, and the optimum temperature for the polymerase used - but PRIMARILY for the binding stringency. DNA and oligonucleotides are the very devil for non-specific binding at low temperatures; and of course ANY binding of a primer to a DNA sequence will set off a reaction
And yes, I have helped develop a number of LAMP tests, now used commercially. Each has its optimum conditions, and even small departures from these (Mg++ concentration, primer sequences and concentrations, dNTP concentrations, reaction temperature, polymerase concentration, you name it) are just begging for non-specific reactions. These things are ALWAYS a balancing act between sensitivity (number of DNA or RNA targets detectable) and specificity,
At their best LAMP-based NAATs are very fast, sensitive, and specific; and need require no expensive readers or thermocyclers, They are also much less sensitive to many of the inhibitors found in clinical samples, and sometimes need no more nucleic acid extraction/purification than a simple boiling step to lyse infected cells and virus particles and release the virus DNA or RNA
Complete reaction mixes can be supplied freeze-dried, needing only to be reconstituted with sample extract in appropriate buffer, and heated. Freeze-dried reaction mixes can be made by the million in beaded form. I know this because I was responsible for the formulation and validation of the first ever freeze-drying formulations for LAMP tests.
A long time ago now, as these things go.